基于支持向量机的遗忘型轻度认知障碍个体识别研究

发布时间：2018-04-23 15:09

本文选题：遗忘型轻度认知障碍 + 灰质体积　；参考：《医学研究生学报》2014年08期

【摘要】：目的近年来多元模式分析(multivariate pattern analysis,MVPA)方法的出现被认为是可以对各种神经精神疾病进行自动化识别的很有前途的工具,支持向量机(support vector machine,SVM)则是一种最广泛使用的MVPA方法。文中采用SVM分类器对遗忘型轻度认知障碍(amnestic mild cognitive impairment,aMCI)患者和无记忆障碍及其他相关疾病者进行MVPA研究,旨在构建具有较高判别能力的个体诊断模型,并从多变量分析的角度来解析aMCI患者的灰质损伤模式。方法采用3.0T磁共振对51例aMCI患者和68例正常对照者进行高分辨率三维T1-weighted扫描,为每个受试者计算灰质体积图谱,该图谱用于之后的判别分析。使用特征选择方法去除冗余信息后训练SVM分类器,使用留一交叉验证估计分类器的性能,最后识别出最有判别能力的灰质模式。结果该方法的分类准确率为83.19%,敏感性为76.47%,特异性为88.24%,接收者操作特性曲线下的面积是0.8368。对分类贡献最大的灰质区域包括双侧海马旁回、双侧海马、双侧杏仁核、双侧丘脑、右侧扣带回、右侧楔前叶、左侧尾状核、左侧颞上回、左侧颞中回、左侧岛叶以及左侧眶额皮层。结论构建的分类模型对aMCI患者具有较好的识别能力,可显示aMCI患者全脑灰质萎缩情况,对临床早期诊断aMCI患者有重要意义。
[Abstract]:Objective in recent years, the multivariate pattern analysis (MVPA) method is considered to be a promising tool for the automatic identification of various neuropsychiatric disorders, and support vector machine (SVM) is the most widely used MVPA method. In this paper, SVM classifier was used to study MVPA in amnestic mild cognitive impairment of amnesia and those without memory disorder and other related diseases. The purpose of this study was to construct a diagnosis model with high discriminant ability. The gray matter damage model of aMCI patients was analyzed by multivariate analysis. Methods High-resolution three-dimensional T1-weighted scanning was performed on 51 patients with aMCI and 68 normal controls with 3.0T magnetic resonance imaging. The gray matter volume map was calculated for each subject and used for later discriminant analysis. The feature selection method is used to train the SVM classifier after the redundant information is removed, and the performance of the classifier is estimated by using a cross-validation method. Finally, the gray matter pattern with the most discriminating ability is identified. Results the classification accuracy of this method is 83.19, the sensitivity is 76.477.The specificity is 88.24.The area under the operating characteristic curve of receiver is 0.8368. The areas of gray matter that contribute most to the classification include bilateral perihippocampal gyrus, bilateral hippocampus, bilateral amygdala, bilateral thalamus, right cingulate gyrus, right anterior cuneate lobe, left caudate nucleus, left superior temporal gyrus, left middle temporal gyrus. Left insular lobe and left orbital frontal cortex. Conclusion the classification model has good recognition ability for aMCI patients and can show the whole brain gray matter atrophy in patients with aMCI. It is of great significance for early diagnosis of aMCI patients.
【作者单位】：首都医科大学宣武医院神经内科;牡丹江医学院红旗医院;北京师范大学认知神经科学与学习国家重点实验室&IDG/麦戈文脑科学研究院;牡丹江医学院红旗医院神经内科;首都医科大学宣武医院放射科;
【基金】：国家自然科学基金(81261120571,30970823,31371007) 北京市科委首都市民健康培育项目(Z131100006813022)
【分类号】：R749.16

【相似文献】